<sub id="ftvvz"></sub>
<address id="ftvvz"></address>
<form id="ftvvz"></form>
<thead id="ftvvz"></thead>

<sub id="ftvvz"></sub>

學術報告
您現在的位置: 首頁 > 科學研究 > 學術報告 > 正文

20201128 劉李 Semi-Supervised Active Learning for COVID-19 Lung Ultrasound Multi-symptom Classification

發布時間:2020-11-23 17:12    瀏覽次數:    來源:

題目:Semi-Supervised Active Learning for COVID-19 Lung Ultrasound Multi-symptom Classification

報告人:劉李 研究科學家(深圳大數據研究院)

時間:2020/11/28 下午16:00-17:00

騰訊會議鏈接:https://meeting.tencent.com/s/EzIyp2VTJUeD

騰訊會議 ID297 688 458

摘要:Firstly, Deep learning and Machine learning will be briefly introduced. Some hot spots on deep learning will be discussed. Then, our recent work on COVID-19 Lung Ultrasound Multi-symptom Classification will be presented. Ultrasound (US) is a non-invasive yet effective medical diagnostic imaging technique for the COVID-19 global pandemic. However, due to complex feature behaviors and expensive annotations of US images, it is difficult to apply Artificial Intelligence (AI) assisting approaches for the lung’s multi-symptom (multi-label) classification. To overcome these difficulties, we propose a novel semi-supervised Two-Stream Active Learning (TSAL) method to model complicated features and reduce labeling costs in an iterative manner. Moreover, a novel lung US dataset named COVID19-LUSMS is built, currently containing 71 clinical patients with 6,836 images sampled from 678 videos. Experimental evaluations show that TSAL can achieve superior performance to the baseline and the state-of-the-art using only 20% data. Qualitatively, visualization of the attention map confirms a good consistency between the model prediction and the clinic knowledge.

個人簡介:劉李,1992年出生,2018年博士畢業于法國的格勒諾布爾-阿爾卑斯大學GIPSA實驗室,之后在Ryerson University大學做博士后研究,后來入選深圳孔雀計劃,現在在深圳大數據研究院工作。是概率論、語音圖像處理,多模態融合,醫療圖像,機器學習及深度學習等領域年輕一代的專家,已經寫出并發表多篇有影響力的論文。

 

湖南大學版權所有?2017年    通訊地址:湖南省長沙市岳麓區麓山南路麓山門     郵編:410082     Email:xiaoban@hnu.edu.cn
域名備案信息:[www.hnu.edu.cn,www.hnu.cn/湘ICP備05000239號]      [hnu.cn 湘教QS3-200503-000481 hnu.edu.cn  湘教QS4-201312-010059]

体育彩票下载